An Innovative Search Algorithm Based On Dynamic Distribution
The Artificial Intelligence (AI) has caught people’s attention since the google software AlphaGo has successfully defeated those best Go players around the world. It is well known that the dynamic programming is the core technics of the Artificial Intelligence. Among the dynamic programming algorithms, the Evolutionary Algorithms are the most common implemented approaches. The notable and popular methods are the Genetic Algorithm, Evolutionary Strategy and Genetic Programming. These algorithms are also referred as simulation optimization. Since most of the real world problems are more complicated than before, an efficient searching optimization algorithm is more important than before. How to find the best solution from other thousands combinations in time is the goal of today’s simulation computation and artificial intelligence. Many other available researches presented the results based on normal distribution, which might take longer computation and lost the solution accuracy. In this research, we propose a dynamic distribution that can take care the diversity and intensity of the solution, and can be utilized in many real world simulations. Simulation results are presented to verify the proposed algorithm.
Keyword: Dynamic Programming, Evolutionary Algorithms